Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# Re: st: marginal after biprobit

 From Ayman Farahat <[email protected]> To "[email protected]" <[email protected]> Subject Re: st: marginal after biprobit Date Thu, 22 Nov 2012 12:26:22 -0800 (PST)

```Thanks Dimitriy
That was very helpful.
I have a follow up question on how to interpret the correlation coefficient (artrho, or rho)

In case when i have :
biprobit (isclosed = myfailpricerisk lreviewnumber) (isgroupon = mygrouponziprisk mygrouponpricerisk lbrate)
The correlation measure the degree of association between the un-observed factors in equation 1 (isclosed) and equation 2 (isgroupon)
so if i have a positive and significant correlation i can conclude that the unobserved share some common  determinants .

In second case
biprobit (isclosed = myfailpricerisk lreviewnumber ISGROUPON) (ISGROUPON = mygrouponziprisk mygrouponpricerisk lbrate)
how do you interpret the correlation?
I have found that the correlation switch signs from first to second case?

Best
Ayman

----- Original Message -----
From: Dimitriy V. Masterov <[email protected]>
To: Statalist <[email protected]>
Cc:
Sent: Wednesday, November 21, 2012 5:03 PM
Subject: Re: st: marginal after biprobit

Ayman,

biprobit is the right command. Austin Nichols has instructions on
calculating the AME here:
www.stata.com/meeting/chicago11/materials/chi11_nichols.pdf.

There are several ways of doing it, some are better than others, but
they tend to produce similar results.

DVM

On Wed, Nov 21, 2012 at 4:47 PM, Ayman Farahat <[email protected]> wrote:
> Hello ;
> I am working on a problem where the dependent variable (iscloed)  is binary and one of the regressors (isgroupon) is binary and endogenous.
> Based on advice i got from the statalist, it was suggested to use a biprobit,
> I ended up using something of the form
>
> biprobit (isclosed = myfailpricerisk lreviewnumber isgroupon) (isgroupon = mygrouponziprisk mygrouponpricerisk lbrate)
>
>
> A couple of questions :
> 1)  Is the biprobit the correct command?
> 2) how do i get the marginal effect of "isgroupon" on "iscolosed" ?
> Thanks
> Ayman
>
> *
> *   For searches and help try:
> *  http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*  http://www.stata.com/help.cgi?search
*  http://www.stata.com/support/faqs/resources/statalist-faq/
*  http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
```